Search

Germline epigenetic programming, including genomic imprinting, substantially influences offspring development. Polycomb Repressive Complex 2 (PRC2) plays an important role in Histone 3 Lysine 27 trimethylation (H3K27me3)-dependent imprinting, loss of which leads to placental hyperplasia in mammalian offspring generated by somatic cell nuclear transfer (SCNT). In this study, we show that offspring from mouse oocytes lacking the Polycomb protein Embryonic Ectoderm Development (EED) were initially growth restricted, characterised by low blastocyst cell counts and substantial mid-gestational developmental delay. This initial developmental delay was followed by striking late-gestational placental hyperplasia, fetal catch-up growth and extended gestational length that culminated in offspring overgrowth. This involved remodelling of the placenta, including expansion of fetal and maternal tissues and conspicuous expansion of the glycogen enriched cell population in the junctional zone that was associated with a delay in parturition. Despite this remodelling and offspring catchup growth, fetal/placental weight ratio and fetal blood glucose levels were low indicating low placental efficiency. Genome-wide analyses identified extensive transcriptional dysregulation in affected placentas, including a range of imprinted and non-imprinted genes and increased expression of the H3K27me3-imprinted geneSlc38a4,which regulates transport of essential amino acids in the placenta. Our data provide an explanation for apparently opposing observations of growth restriction and overgrowth of offspring derived fromEed-nulloocytes and demonstrate that PRC2-dependent programming in the oocyte regulates fetal and placental growth and developmental outcomes.

Background Non-genetic disease inheritance and offspring phenotype are substantially influenced by germline epigenetic programming, including genomic imprinting. Loss of Polycomb Repressive Complex 2 (PRC2) function in oocytes causes non-genetically inherited effects on offspring, including embryonic growth restriction followed by post-natal offspring overgrowth. While PRC2-dependent non-canonical imprinting is likely to contribute, less is known about germline epigenetic programming of non-imprinted genes during oocyte growth. In addition, de novo germline mutations in genes encoding PRC2 lead to overgrowth syndromes in human patients, but the extent to which PRC2 activity is conserved in human oocytes is poorly understood. Results In this study, we identify a discrete period of early oocyte growth during which PRC2 is expressed in mouse growing oocytes. Deletion of Eed during this window led to the de-repression of 343 genes. A high proportion of these were developmental regulators, and the vast majority were not imprinted genes. Many of the de-repressed genes were also marked by the PRC2-dependent epigenetic modification histone 3 lysine 27 trimethylation (H3K27me3) in primary–secondary mouse oocytes, at a time concurrent with PRC2 expression. In addition, we found H3K27me3 was also enriched on many of these genes by the germinal vesicle (GV) stage in human oocytes, strongly indicating that this PRC2 function is conserved in the human germline. However, while the 343 genes were de-repressed in mouse oocytes lacking EED, they were not de-repressed in pre-implantation embryos and lost H3K27me3 during pre-implantation development. This implies that H3K27me3 is a transient feature that represses a wide range of genes in oocytes. Conclusions Together, these data indicate that EED has spatially and temporally distinct functions in the female germline to repress a wide range of developmentally important genes and that this activity is conserved in the mouse and human germlines.